Determinants of Default Probabilities:
A BMA Approach

Abstract

There is no consensus about the determinants of corporate defaults. In this paper, we employ Bayesian Model Averaging (BMA) techniques in order to address this issue. This is a novel approach in the credit risk literature. We consider a comprehensive set of candidate regressors including macroeconomic, sectorial, and firm-specific (both accounting and market) variables. The most conclusive evidence is for the firm stock return and the ratio of working capital to total assets. However, we also find that many other variables such as the industrial production index growth rate, the ratio of retained earnings to total assets, and the firm’s leverage can also be considered as robust determinants of corporate default.

JEL Classification: .
Keywords: .